Suspension-tomsurface Heat Transfer in a Circulating-Fluidized-Bed Combustor Heat transfer coefficients were measured for circulating beds of sand particles of mean size 222 to 299 pm at temperatures of 340-880°C. Transfer coefficients were obtained for both a 1.22-m-long, 12.7-mm- OD vertical tube and a 1.59-m-long, 148-mm-wide membrane wall near the top of a 152-mm-square by 7.32-m-tall combustion column. For both surfaces and all temperatures, average heat transfer coefficients increased almost linearly with local suspension density which ranged from 0 to 70 kg/m3. Radiation played a significant role, especially at high temperatures and low suspension densities. Heat transfer coeffi- cients also varied significantly with the lateral position of the tube. The vertical length of heat transfer surface is shown to be an important parameter allowing seemingly discrepant published results to be recon- ciled. Richard L. Wu John R. Grace C. Jim Lim Clive M.H. Brereton Department of Chemical Engineering and Pulp and Paper Center University of British Columbia Vancouver, Canada V6T 1 W5 Introduction While circulating fluidized beds continue to gain wide accep- tance, especially for gas-solid reactions like calcination and combustion, there are still many fundamental areas which are little understood. Research in heat transfer in circulating beds, for instance, remains largely empirical with few published results available for public dissemination (Grace, 1986; Glicks- man, 1988). Theoretical modeling attempts are hindered by the fact that these published data are extremely scattered when compared against one another. In this paper, we present experimental data for sand particles of 222 to 299 wm obtained from a tube and a membrane wall at temperatures of 34O-88O0C. The effect of lateral position of the tube is also investigated. Analyses of local heat transfer coeffi- cient profiles along the membrane wall are then presented to demonstrate the significant influence of the vertical length of heat transfer surface. Experimental Equipment All the experimental data in this paper were obtained using the circulating-fluidized-bed combustion (CFBC) facility at the University of British Columbia. Many details of the equipment are provided in previous papers (Wu et al., 1987; Grace et al., 1987; Legros et al., 1989). Only a brief description of the key details relevant to the heat transfer results are provided here. A schematic of the major experimental components is shown in Figure I. The refractory-lined reactor column is 7.32-m-high and 152 x 152 mm in cross section. For the studies with the membrane transfer surface, the bottom section of the reactor column was a refractory-lined stainless-steel section with its inside tapered from a 51 x 152 mm cross section to 152 x 152 mm cross section over its 1.22-m height to provide a high accel- eration zone, reducing sintering and agglomeration. The entire riser is instrumented with thermocouples and pressure taps at 0.61 m intervals along opposite walls for the determination of temperature and density profiles. For the tapered bottom sec- tion, primary air was introduced through a horizontal half-pipe cut along its axis to form a distributor with twenty 9.5-mm- diameter orifices drilled on its curved surface. For the experi- ments with the vertical cooling tube, the distributor had three tuyeres each with six inclined holes. Preheated secondary air was introduced through two pairs of directly opposed air ports 0.9 m above the distributor. Solids entrained in the column were continuously captured by a 0.3 1-m-ID refractory-lined primary cyclone and returned 0.4 m above the bottom of the column via an L-valve. The solids recirculation rate was controlled by vary- ing the flow of aeration air to the L-valve. Fines not captured by the primary cyclone were captured with a 0.2-m-ID secondary cyclone and could be returned to the bottom of the reactor via an eductor. The first set of heat transfer data was obtained from a 1.22- m-long, 12.7-mm-OD water-cooled stainless-steel tube begin- ning 4.57 m above the distributor on one wall of the column. This tube was normally positioned touching the refractory sur- face midway between two faces of the column. It could also be moved to the axis of the column or intermediate lateral posi- AIChE Journal October 1989 Vol. 35, No. 10 1685